How To conduct a thesis 1- Define the problem 2- Reviewing the relevant literatures 3- Preparing a protocol :the contents of the protocol: Description (abstract) and personnel Preliminary studies Research plan: 1. Specific objectives 2. Backgrounds and significance

Methods: Overview of design (time frame and nature of control) Study subjects (specify the target and accessible population and design for sampling Measurements :*main predictor variables (intervention) *Potential confounding variable. *Outcome variable Pretest plan Statistical issues Quality control and data management Time table and organizational chart - Ethical considerations (consultants and arrangements between institutions)

How to write a thesis

General structure: Introduction (what question was asked?) Methods (how was it studied?) Results (what was found?) Discussion (what do the findings mean?)

The cover page The title Description Name (and qualification) of the student Name (and qualification) of the supervisor Year Arabic cover page (and others)

The title Should be: - As short as possible Simple and in easy understood words Represent the research question (objective) Convey what the paper is about Draw the attention of the readers Interesting and exciting Be alert that it matches the conclusion

Dedication

Acknowledgement

Abstract Abstract and summary Structured and unstructured abstract It should describe succinctly: - Why what was done was done - What was done What was found What was concluded

Abstract Structured: Background Objective(s) Methods Results Conclusion

Abstract It should: Contain the essence of the whole paper Stand alone Stick to a maximum of 150 words for the unstructured abstract and 250 for the structured Be clear, concise, readable and informative Avoid unnecessary details Avoid abbreviations Do not mention references

Lists: of contents of abbreviations (if applicable) of tables

The introduction should be: - Brief and concise (between 7-10 pages) - State clearly the research question - Give an idea (background) about the subject - Call attention to the importance of the issue - Include the most interesting details - Contain up to date information - Specially rich with facts and numbers - Cited from reliable references - Lead the reader gradually to the objective(s) of the study (at the end of the introduction).

Methods Settings Study design The sample (who, what, where when, and how) Sample size The sampling technique Give accurate details of materials used Detailed procedures Give exact form of treatment and drug dosage (if any) The questionnaire (if applicable) Statistics Test used Program used P value, Beta error,…

Methods Other important points: Ethical issue Control and comparison groups Selected groups - Random and systematic errors - Blindness - Approval and consents Financial support Limitations of the study

A good methods section can answer these questions: - Does the text describe what question was being asked? - Does it describe what was being tested? What is the degree of accuracy and precision of the measurements used for the variables? Where these trustworthy measurements recorded, analyzed, and interpreted correctly? - Would a suitably qualified reader be able to repeat the experiment in the same way?

The Results - Tell the story of how you arrived at the answers. Establish initially how normal or abnormal your groups were and how comparable they were - Even with a random allocation into groups; it is necessary to confirm that they are equivalent - Mention if your sample was homogenous - Describe the main characteristics of your sample - Don’t be afraid of negative results - State the statistical significance Do not over-regard your results

The statistics - Use the right test - Set the dependent and independent variables - Set the P value - Do not give results to a greater degree of accuracy than that of the measurement Do not use percentages unless the groups contain more than 100 subjects - Be aware about the type of association - Be cautious about mentioning a (causal) relationship Consider the statistical power (sample size, randomization, No. of end-points,..) - Always mention the confidence interval

The tables Should be: Clear, well organized and neat of errors Simple and not complicated Self explanatory The title should be put above the table The title should answer the questions: what, who, where, and when Put in the right direction (dependent and independent variables) Mention the totals (and percentages if necessary) Mention the reference (if any) Mention the statistical values (Test of significance, P value, C.I., df,..) Better not to be too many

Illustrations The mind takes in pictorial information more quickly than written text Show the most significant results Make it in colors Make it simple and detailed For the audience (presentation) not for the thesis

Photographs, micrographs and graphics Photographs must include means to protect the anonymity of any patients Micrographs needs measures of scale Graphics: three dimensional representations (Bar graphs and pie charts) are irresistible. But use it as a last resort to bring out a point. All need professional production

Discussion The main findings (interpretation) Previous work (comparison) Discussion of methods What it means for practice (implications) The need for further studies Bias and Limitations Conclusions and recommendations

References Harvard Vancouver

Other important issues The Authors The printing The language Approvals Questionnaire Arabic title and abstract Others